Finally, it works! After two months of coding around the issues with PyLucene from ports I finally tried (upon Andi's suggestion to me last month) compiling PyLucene with gcj 3.4.6. To my surprise, all the make tests have passed and I have had no issues thus far.
To get it running, the Makefile for the gcc-3.4.6 port has to be patched as well as PyLucene's Makefile. Later, I'll submit a page to the wiki documenting the whole process with the relevant patches. In the mean time if anybody needs some help, I'll be glad to help. Thanks Andi! TJ Ninneman
--- Begin Message ---On Mon, 28 May 2007, Krys Wilken wrote: > Just wanted to let you know that I have compiled Pylucene 2.1.0-2 for Python > 2.5 on Win32 and made it availalbe at: Great. Which version of gcj did you use ? > - testCategorization: Python cannot "float('nan')" on Windows. This is > platform-specific. I don't think this is a PyLucene error. This is > relevant, I think: Yeah, a known problem. > - testTiming: Compound indexing time is a bit faster than multi-file indexing > time. I don't know if this is a big deal or not, but I suspect it is not. That one too. It varies and sometimes fails because the expected faster test is slower. It usually passes. Andi.. _______________________________________________ pylucene-dev mailing list [email protected] http://lists.osafoundation.org/mailman/listinfo/pylucene-dev
--- End Message ---
_______________________________________________ pylucene-dev mailing list [email protected] http://lists.osafoundation.org/mailman/listinfo/pylucene-dev
